[QE-users] HSE calculation for ZnO band structure converges slow
Stefano Baroni
baroni at sissa.it
Wed Aug 15 08:06:37 CEST 2018
Which version of QE are you using? SB
___
Stefano Baroni, Trieste -- http://stefano.baroni.me
> On 15 Aug 2018, at 04:43, Tan Hengxin <tanhx90 at gmail.com> wrote:
>
> Dear Users,
>
> I am doing HSE calculations on wurtzite ZnO for the band structure.
> However, the run converges slowly:
> I use a 6*6*4 k-mesh. The cutoff energy is 60 Ry. Norm-conversing pseudopotentials are used, and there are 12 and 6 valence electrons for Zn and O respectively.
> The job was run with 2 nodes (48 processors) with npool = 2. The run takes 45 hours.
>
> Is this seems normal? Or what would be done to reduce the run time? Are there any tricks that need to be paid special attention to ZnO?
> Thanks for your help.
>
> The input parameters and the tail of the output are copied below.
> input:
> &CONTROL
> calculation = 'scf',
> prefix = 'ZnO',
> pseudo_dir = './',
> verbosity = 'high',
> wf_collect = .true.
> etot_conv_thr = 1.0D-6,
> forc_conv_thr = 1.0D-4,
> restart_mode = 'from_scratch',
> outdir = './temp_out',
> /
> &SYSTEM
> ibrav = 0,
> nat = 4,
> ntyp = 2,
> ecutwfc = 60,
> nbnd = 36,
> input_dft = 'hse',
> exx_fraction = 0.25,
> nqx1 = 6, nqx2 = 6, nqx3 = 6,
> /
> &ELECTRONS
> mixing_mode = 'plain',
> mixing_beta = 0.7,
> conv_thr = 1.D-8,
> /
>
> ATOMIC_SPECIES
> Zn 65.38 Zn_fan_nc_pbe_srl.upf
> O 15.999 O_web_nc_pbe.upf
>
> CELL_PARAMETERS (angstrom)
> 3.249 0.000 0.000
> -1.625 2.814 0.000
> 0.000 0.000 5.205
>
> ATOMIC_POSITIONS (crystal)
> Zn 0.333333343 0.666666687 0.000000000
> Zn 0.666666627 0.333333313 0.500000000
> O 0.333333343 0.666666687 0.382600009
> O 0.666666627 0.333333313 0.882600009
>
> K_POINTS {crystal}
> 216
> (The 216 k points with weights (from 6*6*6 k-mesh))
> output:
> ......
>
> ! total energy = -269.06624688 Ry
> Harris-Foulkes estimate = -277.03747208 Ry
> estimated scf accuracy < 3.7E-09 Ry
>
> convergence has been achieved in 1 iterations
>
> ! total energy = -269.06624688 Ry
> Harris-Foulkes estimate = -269.06624688 Ry
> est. exchange err (dexx) = 0.00000000 Ry
> - averaged Fock potential = 15.94246020 Ry
> + Fock energy = -7.97123500 Ry
>
> EXX self-consistency reached
>
> Writing output data file ZnO.save
>
> init_run : 1.05s CPU 1.15s WALL ( 1 calls)
> electrons : 150333.27s CPU 152710.02s WALL ( 5 calls)
>
> Called by init_run:
> wfcinit : 1.01s CPU 1.11s WALL ( 1 calls)
> wfcinit:atom : 0.00s CPU 0.00s WALL ( 108 calls)
> wfcinit:wfcr : 1.00s CPU 1.03s WALL ( 108 calls)
> potinit : 0.01s CPU 0.02s WALL ( 1 calls)
>
> Called by electrons:
> c_bands : 150323.40s CPU 152695.15s WALL ( 25 calls)
> sum_band : 7.77s CPU 8.01s WALL ( 25 calls)
> v_of_rho : 0.12s CPU 0.13s WALL ( 27 calls)
> v_h : 0.00s CPU 0.00s WALL ( 27 calls)
> v_xc : 0.12s CPU 0.12s WALL ( 27 calls)
> mix_rho : 0.01s CPU 0.01s WALL ( 25 calls)
>
> Called by c_bands:
> init_us_2 : 0.21s CPU 0.24s WALL ( 6480 calls)
> cegterg : 149815.07s CPU 152046.20s WALL ( 2700 calls)
>
> Called by sum_band:
>
> Called by *egterg:
> h_psi : 149780.99s CPU 152011.56s WALL ( 9389 calls)
> g_psi : 0.08s CPU 0.13s WALL ( 6581 calls)
> cdiaghg : 29.34s CPU 29.97s WALL ( 8849 calls)
> cegterg:over : 1.80s CPU 1.79s WALL ( 6581 calls)
> cegterg:upda : 1.33s CPU 1.31s WALL ( 6581 calls)
> cegterg:last : 0.86s CPU 0.84s WALL ( 2808 calls)
> cdiaghg:chol : 1.34s CPU 1.48s WALL ( 8849 calls)
> cdiaghg:inve : 0.96s CPU 0.98s WALL ( 8849 calls)
> cdiaghg:para : 1.74s CPU 1.89s WALL ( 17698 calls)
>
> Called by h_psi:
> h_psi:pot : 39.93s CPU 40.82s WALL ( 9389 calls)
> h_psi:calbec : 1.34s CPU 1.29s WALL ( 9389 calls)
> vloc_psi : 38.33s CPU 39.21s WALL ( 9389 calls)
> add_vuspsi : 0.23s CPU 0.28s WALL ( 9389 calls)
>
> General routines
> calbec : 30.41s CPU 30.38s WALL ( 10361 calls)
> fft : 0.08s CPU 0.09s WALL ( 289 calls)
> fftw : 44.34s CPU 44.67s WALL ( 579174 calls)
> fftc : 136683.94s CPU 138542.75s WALL (******** calls)
> fftcw : 26.63s CPU 26.25s WALL ( 339050 calls)
>
> Parallel routines
> fft_scatter : 82451.93s CPU 77795.40s WALL (******** calls)
>
> EXX routines
> exx_grid : 0.10s CPU 0.10s WALL ( 1 calls)
> exxinit : 38.19s CPU 176.06s WALL ( 5 calls)
> vexx : 149740.98s CPU 151970.61s WALL ( 5822 calls)
> exxenergy : 8652.21s CPU 8774.70s WALL ( 9 calls)
>
> PWSCF : 1d 20h10m CPU 1d 20h56m WALL
>
>
> This run was terminated on: 9:57:17 10Jul2017
>
> =------------------------------------------------------------------------------=
> JOB DONE.
> =------------------------------------------------------------------------------=
>
> Tan, Hengxin
> Department of physics, THU.
> Beijing 100084, China
> Office: B403,New Science Building
> E-mail:tanhx90 at gmail.com
> _______________________________________________
> users mailing list
> users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20180815/2853898a/attachment.html>
More information about the users
mailing list